EFFICIENCY OF INTEGRATED SEISMIC METHODS APPROACH TO NEAR-SURFACE CHARACTERIZATION

Authors:

Irena GJORGJESKA1*, Vlatko SESOV1, Kemal EDIP1, Dragi DOJCINOVSKI1

1Institute of Earthquake Engineering and Engineering Seismology (IZIIS), Ss. Cyril and Methodius University in Skopje, Republic of North Macedonia

*email: gj_irena@iziis.ukim.edu.mk

Geophysical survey by use of combined active seismic methods at three characteristic locations in R. North

Macedonia were performed

Clinical Center, SkopjeSeismic Refraction vs 2D MASW

Konsko, GevgelijaSeismic Refraction vs 1D MASW

Kurshumli Ann SkopjeSeismic Refraction vs Reflection & 2D MASW

Locations and applied methods

The main objective of this study is to show the advantages and efficiency of using an integrated seismic methods approach to subsurface modeling.

Fig.1. Regional and local maps of the study locations

Location Coordinates (UTM WGS 84)

Kurshumli Ann, Skopje 7 536 195.00 E 4 650 211.00 N

Clinical Center, Skopje 7 534 817.00 E 4 648 702.00 N

Konsko, Gevgelija 7 611 555.03 E 4 559 616.76 N

• The same seismic equipment and, in most of the cases, the same acquisition parameters were used

for the surveys

• The measurements were performed using the SoilSpy Rosina multichannel digital seismograph (MoHo - Science & Technology, Italy).

• The seismic energy was generated with vertical impacts by a 10 kg sledge hammer on an aluminum plate and was recorded by 4.5 Hz vertical geophones.

Fig.2. Photos of the survey locations

Software and methodology

Seismic refraction & reflection data processing was performed using

ReflexW software and MASW data using SurfSeis 3.06 software

Seismic refraction tomographicconcept

• Based on a gridded initial model for theiterative process, to determine the velocityof individual 2-dimension grids within aprofile

• Opposed to modeling the subsurfacestructure as constant velocity layers so-called “cake layers”

Seismic reflection

Processing of the seismic reflection data was performed using the

Common Mid Point (CMP) technique. Pre-stacking static

correction and 1D filtering was applied on the raw data as was also

post-stacking 2D filtering and depth migration.

MASW

• The dispersion curves were extracted for each source-receivers configuration displacement.

• The inversion was performed for each of the dispersion curves by the iterative process

Location: Clinical center, Skopje

• Preliminary MASW surveys were first conducted to optimize the field acquisition parameters, such as the offset of the seismic array, the receiver spacing and the length of the record

• The surveys were performed varying the acquisition parameters in order to determine their respective influence on the dispersion image resolution

a) 1st source-receiver configuration displacement

b) 2nd source-receiver configuration displacement

c) 1st source-receiver configuration displacement

d) 2nd source-receiver configuration displacement

Fig3. Dispersion images with extracted effective dispersion curves (white dots)

a) and b) 1st seismic array design: 16 channels, distance between geophones of 3m, minimum off-set of 15m and recording length duration of 1.5s, with a sampling frequency of 1024 Hz, 11 SR displacement in total, 3m excitation step

c) and d) 2nd seismic array design: 16 channels distance between geophones of 2m, minimum off-set of 5m and recording length duration of 0.3s, with a sampling frequency of 1024 Hz, 19 SR displacement in total, 2m excitation step

Fig.4 2D Vs model – Multichannel Analysis of Surface Waves (MASW). Generated using the data from the 2nd seismic array design.

Fig.5. Seismic refraction tomography 2D Vp and Vs model. The surveyswere conducted along 17 channels seismic spread with receiver distanceof 5m, near offset of 5m and recording length of 0.5s

• Data analysis with the 2nd configuration ofseismic array with 5m minimum offset and0.3s recorded signal length(unconventional for this kind of surveys)provided a better quality dispersion image.

• The effective dispersion curves wereextracted from the dispersion images as acombination of the fundamental and highermodes of the Rayleigh waves.

• The generated 2D MASW Vs model (Fig.4)is a result from the survey using 2nd seismicarray design.

• According to the results from MASW andseismic refraction survey, the surface layersof the terrain are characterized by seismicvelocities in the range of Vp=350-1700 m/s,and Vs=130-620m/s

• They are composed of quaternary deposits,overlying Mio-Pliocene sediments,characterized by Vp>1800m/s, Vs>680m/s.

• Both, the seismic refraction and MASW 2Dmodel clearly mapped the seismic bedrocktopography i.e the max. thickness of thequarternary deposits in this part of thelocation is approximately 10-15 m

• The velocity inversion is mapped at 7 m inthe 2D Vs MASW model and indicates agroundwater level

Survey design and acquisition parameters

Seismic refraction Seismic reflection and 2D MASW

• 17 channels• 3m interstation distance• 3m minimum offset• 0.5s seismic record • 1024 Hz sampling

frequency • excitation step 11m

Roll-a-long mode

• 16 channels• 2m interstation distance• 6m minimum offset• 0.5s seismic record • 1024Hz sampling frequency• 13 source-receiver spread displ.• excitation step 2m

Location: Kurshumli Ann, Skopje

• The surface layers of the terrain are composed of quaternary, alluvial-proluvial deposits, characterized by seismic velocities in the range ofVp=170-1750 m/s, and Vs=100-630m/s.

• They overlying Pliocene sediments which are mainly composed of gravel, sand, sandstones etc., characterized by Vp>1800m/s, Vs>680m/s• The thickness of the quaternary deposits varies in the range of 8m to 15m. The anomaly is clearly mapped on the seismic refraction model (along a

distance of 10-38m) and the MASW model.• The same variation of the seismic bedrock topography is interpreted at the seismic reflection 2D model. The reflection model indicates

deformations in the deeper layers, as well.• The velocity inversion mapped at 4-5m in the 2D Vs MASW model indicates a groundwater level.

Fig.6. 2D model as a result of the survey performed in Skopje, Kurshumli Ann a) Vp seismic refraction tomography model b) 2D Seismic reflection section. c) 2D Vs model as a result of the MASW survey

Location Konsko, Gevgelija:

Survey design and acquisition parameters

Seismic refraction 1D MASW

2 Seismic profiles

• Rp7: 34 channels• Rp11: 17 channels• 5m interstation distance• 5m minimum offset• 0.5s seismic record • 256 Hz sampling frequency• 1024Hz sampl. frequancy

3 Seismic profiles

• 17 channels• 5m interstation distance• 5m minimum offset• 0.5s seismic record • 1024Hz sampl. frequancy

• The terrain of the location is composed of gabro covered with diluvial material.

• The main objective of the survey was definition of the depth of the surface critical zone: diluvial material with clay infill and layer of intensively cracked, degraded rocks (Vs=100-480m/s )

Fig.7. 2D Vp seismic refraction model Rp7 (Fig1)as a result of the survey performed in Konsko

Fig.8. a-1) Dispersion image D1 with extracted dispersion curve (white dots). 1D MASW survey along first half of Rp7 seismic profile. a-2) 1D Vs model as a result of D1 dispersion curve inversion (refers to 40m position of the Rp7 profile). b-1) Dispersion image D2 with

extracted dispersion curve (white dots). 1D MASW survey along second half of Rp7 seismic profile. b-2) 1D Vs model as a result of D2 dispersion curve inversion (refers to 120m position of the Rp7 profile)

• The high impedance contrast between the surface degraded layers and more compact rock layers in their base contributed to extraction of good quality dispersion curves.

• According to the seismic refraction models, the max.thickness in this part of the location is approximately 20-22m

• The reliability of the results is confirmed by the 1D Vs MASW models

Fig.8. 2D Vp seismic refraction model Rp11 (Fig1)as a result of the survey performed in Konsko

Fig.9. a) Dispersion image D3 with extracted dispersion curve (white dots). 1D MASW surveys along Rp11 seismic profile. b) 1D Vs model as a result of D3 dispersion curve inversion (refers to 40m position of the Rp11 profile)

Each of the techniques showed some limitations and disadvantages, but their application in an integrated approach enabled the results to be compared and to complement each other, which reduced the error likelihood in interpretation.

The in-situ measurements and data processing were conducted in the most practical, cost and time-effectiveway, with the same equipment, and in some cases the same acquisition parameters.

From the above presented can be concluded that using an integrated geophysical approach is very significant for a high quality, accurate and reliable subsurface modeling

Conclusions

The seismic refraction tomographic approach enabled modeling of the subsurface with both lateral and verticalvelocity gradients, which provided high resolution imaging of the subsurface structure and proved to be greattool for subsurface characterization and detecting potential anomalies.

2D MASW survey complemented and improved the subsurface modeling of the investigated location mappingthe velocity inversion i.e. trapped low velocity layer. The roll-a-long technique enabled the data to be used forseismic reflection processing. Using the CMP method for reflection processing resulted in very accurate, highresolution modeling of the subsurface up to the depth of 100m.

The seismic refraction and 1D MASW survey at the Konsko location, performed along the same profile linesusing the same acquisition parameters, proved to be an excellent combination for fast and accurate subsurfacemodeling especially in hard terrain conditions.

Slide Number 1Slide Number 2Slide Number 3Slide Number 4Slide Number 5Slide Number 6Slide Number 7Slide Number 8Slide Number 9Slide Number 10Slide Number 11Slide Number 12Conclusions